Paper Press Rewinder

rsdoran

Lifetime Supporting Member
Join Date
Apr 2002
Location
Birmingham, AL
Posts
7,371
Note: This will probably be another of my long rambles. As most know I am not a "programmer" even though I deal with plc's consistently.

We have a rewinder that is approximately 12 years old using a Klockner-Moeller PS316 plc. The ANALOG cards are not available for this device anymore...IF YOU HAVE ANY LET ME KNOW IF WE CAN OBTAIN THEM, PLEASE.

What has happened is one of the Analog cards has failed, the module is good but the card is bad. We have called in a tech from the company that originally designed the control system, we have "rigged" an alternative to the analog cards. NOTE: I do not like this alternative (I could have done this but didnt think prudent nor liked the idea but y'all know engineers and management).

THis system is unique in that because of the drives (Amicon, which is now GE labeled drives) it needs 2 computers to interface FIX (intellution) and the plc/interface.
OVERVIEW
Rewinder for 30" to 40 " wide paper.
Has 2 spindles/shafts to wind paper on that turret (rotate) then match speed for auto splice sequence, existing drives are AMICON
Uses 2 75HP DC drives for the spindles
Turret motor is smaller and fixed speed...maybe 10HP AC (will have to verify).
PLC is Klockner-Moeller PS316 with 3 16 point Digital Input cards, 2 digital output 16 point, 1 4 channel analog input, 2 4 channel output cards and 1 HSC that I think is dual channel
Interface is Intellution FIX which has to connect to another computer to communicate with the drives.

MY QUESTION is thricefold:
First because this PLC has obsolete parts it is my major concern I need to replace it...I have in stock MANY GE FANUC 90-30 with all the parts needed but the analog cards. NOTE: I would prefer AB and you may understand part of that with the next statement but it may not matter.
Next are the drives, they are older etc and I have 2 Reliance Flexpack 3000 drives in a rewinder cabinet (from a dismantled system) that I would prefer to use...ie the reason for using an AB plc.
I would like to eliminate the computers and use an HMI as the interface because the data that has to be changed is primarily the analog values and if I can interface the GE Fanuc (or an AB plc) directly with the DC Drives I shouldnt have the need.


Questions
1. Does the above seem viable with todays technology to incorporate..ie will the GE Fanuc work or should I go for using an AB PLC?
2. I am not a fool, this is more than I can do in one sitting...ie setup drive and program plc/interface...can I reasonably expect to keep the cost under $15000 US (hopefully $10000 US) even if I have to hire an integrator/programmmer(preferably a programmer that can work with us). NOTE: We have the crew to do the preliminary install/wiring (mainly me) but a crew of hard wire electricians. Technically a programmer is all I think I need, the rest we can easily do.
3. TIME...as usual is of the essence. Is it viable to expect to do this within 3 days of actual downtime (I understand the work may take longer doing the install/wiring/programming).

I know I tend to ramble on but this is a real problem that I have now...the alternative they have come up with is to go get an OLD OLD rewind from a closed factory ...personally I dont think that option is better/cheaper/uses less time than doing an upgrade PLUS it means I MUST deal with more OLD ****. Getting the OLD unit requires sending a crew to disconnect it then ship it, then set it, and reconnect it. Geezus I am OLD and have to deal with myself every day...I get tired of dealing with OLD.

Last but not least...I MUST have FULL access to all programming.

Thanks for listening/reading...all thoughts will be appreciated.
 
Last edited:
Ron, You have your hands full here, but you know that. I cant help to much with the integration of all these components like these other guys can, but I can help you with project management, so maybe questions 2 & 3. #2 - If you have to bring in an outside source for programming, engineering, etc. you better plan on much more than $15,000. I've had to sign off on bills for service techs that were higher for just a few days, eng. is out of this world. #3 - This is probably the biggest factor against you doing this yourself. I've read your posts for a long time now & know you are a very talented tech. However, I undertook a major project involving the integration of a SLC, drives, all the associated wiring, startup, debug, etc a few years ago, (like yourself, this isnt really what I'm supposed to do). I started about 3 weeks before the project writing the program. The install took about 9 days with 6 people working about 15 hour days. I'm know this was on a much larger scale than what your talking about, but if you put it in perspective, its still a large undertaking. My advice, go back to the management & tell them to do it right you need more time, maybe money, then do the whole conversion the best way you see fit. My guess is they are going to trust your judgement. I know this sounds like ****, but its better than not making your deadline & spending more than was budgeted.
 
My first guess is that the 90-30 has plenty of horsepower to do what you need. I've used a 90-30 on a centerwind application in the past. There doesn't look like anything in the Klockner-Moeller that's not also available on either the GE or the SLC platform.

The fact that AB owns Reliance doesn't guarantee that the drives and an AB PLC will mesh seamlessly. PLC and drives are still two different divisions, so there is ample opportunity for finger pointing if the retrofit doesn't go as planned.

Replacing the computer running the FIX application is feasible as long as you're not using the FIX computer to store any production data.

I agree with Mike that your cost estimates for ouside help look low.
 
If you are going to use the Field Bus communication to contol the Reliance drive then AB is more advantage over GE. But If the Reliance drive is to be control thru its I/O card then AB is not advantage over GE.
 
Gee Ron, sounds like an opportune time to further your PLC programming knowledge. Think of it this way, you have a complete system that DID work (and hopefully, worked well), and basically all you're doing is "copying" it, using different components. Be happy that you're NOT designing it from scratch! At least you know that if you duplicate what's there, it WILL work. IOW, someone else has already done the design, which is 1/2 the battle (sometimes more, sometimes less)... :D

I'm pretty sure there are no "tricks" that the K-M PLC could do that a 90-30 can't duplicate! Since you still have access to the original designers, I assume that you have a hard-copy of the current program in the PLC. Therefore, all that would be required would be to translate the K-M code to it's GE equivalent. I myself have no experience with K-M PLCs, but I'm sure as a group we could figure out any "weird" instruction you might come across.

You may want to break up the project into more manageable chunks. Don't overwhelm yourself with the whole thing at once. Especially since you have to minimize downtime. The main concern right now is that the PLC needs to be replaced, so make that project #1. Keeping your "wish list" (new drives, no PCs) in mind, you can prepare for these second, third, etc. projects during the PLC replacement.

I honestly think that the PLC swap should be fairly straightforward, so if you do your homework, you should have very few problems doing it yourself. And there's nothing like a deadline to speed up the learning process!... :p

beerchug

-Eric
 
We are smack dab in the middle of a drive replacement project on a web handling machine using the flex pac/web pack 3000 drives with an AB conrtroller. The web pac drive is a better fit for winders cause it has the ratio detector, among other things, built it. The alternitave is to put the ratio detector in the controller. It sounds like you already own two flexpacs, so it's really what you have to use, I guess.

Our new system uses the controllogix platform and talks to the drives on controlnet, although there are different comms options available, I'm sure.

The controllogix platform is a better fit for drive applications because it has built in "control block" functionality. This is an import from the Reliance automax and DCS systems. Not that a SLC or PLC can't do it, but block was designed for coordinated drive applications with functions like PI, scale, lead, lag, summers, etc.

On the other hand, if I had to do the drives myself, I would have probably gone the PLC route, just to eliminate the controllogix learning curve. As it is, Rockwell did all the drive stuff and all I have to do is interface controllogix to the PLC's and SLC's controlling the front of the machine.

Anyhow, that's my 2 cents. Sounds like an interesting and educational project--have fun, and good luck--keep us posted.
 
Ron:

Why get rid of The Fix? This seems to me to be the easiest of the devices to convert. All you have to do is expart the datbase to ASCII (easy) to make a list of all the tags and their K-M adresses.
Keep careful track of where that same data point winds up in the new PLC. Do a search-and-replace (using your favorite editor) to change the address and driver, inport the database and you're done.

I know that the usual recommendation when doing an changeover is to redo the PLC code from scratch. But I've found, unless you REALLY understand what the machine is SUPPOSED to do (regardless of how it does it) that it's easier to just do a straight traslation of the code (even if it means doing "dumb" things. Example: Some PLCs don't have a specialized "greater than" or "less than" compare function, but the subtraction block serves the same purpose. So, even though you are 99% sure that thre is only a compare going on, I'll still do he subtraction, and then have a branch below to do the compare, rather than just doing the compare. It can make the code UGLY), but if by not changing ANY functionality, you have a machine that works the same way as before.

In thory, you can always clean up the code later. Often there are things you discover in the code about how the machine ACTUALLY works that you didn't know before, and would not have included if you hadn't done a straight translation.

And again, the straight translation will make the conversion of the FIX all that much easier.

As for the 3-day cutover - here's another idea. Get some Phoenix connectors (they make them in 16-plug/socket sizes) and put them into your existing I/O wiring, right before the PLC. When you are ready to test the new PLC, you can just unplug the I/O from the K-M, plug them into the new PLC, and test the heck out of it for 3 days. If, at the end of the three days, you aren't ready, just plug the I/O back into the K-M. In 10 minutes, you are ready to go back into production using the old (realiable?) system. Do more analysis/debug until your next downtime window. Don't start tearing out the K-M PLC until you know you've got the software 100%.

The connectors are an added expense, but the GUARENTEED no-downtime loss-of-production insurance is worth it.
 
Thanks again for the replies...FYI I dont have the existing code and the existing plc does not allow you to extract the file to view. I have just enough software and code to reload the machine if I need to.

As for the FIX, its not doing anything but acting as an HMI, no scada etc. Plus I dont see a need for a computer and this thing has 2, just more to give me problems.

The plan at this time is to get another rewind off an old press at a closed plant...I dont see the cost advantage but it was an upper management decision. This rewind is to be rebuilt but all the details about when and HOW are still open.

At this time we are getting the existing rewind to run...lost an encoder today that is a must so have to wait till tomorrow to start it again. Lost a card for the drive yesterday...so this thing is falling apart faster than we can make it go. The tech brought in has gone to airport now to pick up the drive card, then hopefully he will have time to finish before 6...doubt it though.

Anyway it looks like I have more time to look at all options and see what I feel comfortable doing on my end and what I may need from outside sources.
 
Analog alternative

Ron,
I don't know what alternative method you have to replace the analog
input card, but here is what I would do.
Replace the defective analog input card with a digital input card. Take a Micrologix with an analog input and program it to convert analog in to binary/BCD out.
Connect Micrologix outputs to Klockner Moeller inputs.
Re-program Klockner Moeller PLC to accomodate the
above change.
Take the defective Analog input card and have it fixed
by a third party repair shop (it could be AB they are actually very good at this).
A 12 year old design probably does not use the surface mount technology and so it can probably be done.
Good luck
 
rsdoran said:
FYI I dont have the existing code and the existing plc does not allow you to extract the file to view. I have just enough software and code to reload the machine if I need to.

If the OEM won't provide you with the code to fix a problem THEY produced, then THEY should be covering ALL costs involved with the repair of the machine... :p

I realize that it's out of warranty, but it's not your fault that the PLC is now obsolete... It was their choice to use K-M, no? They're still in business, so they should be able to sell you spare parts (like an analog card) for their machine. Since they can no longer obtain this spare part for their machine, THEY should be the ones finding a way around this problem. Yes, they have "rigged" an alternative, but are they currently working on a REAL fix?

Saying "Sorry, we can't help you" is just not good business. If they can't provide you with a decent "fix", then they should at least provide you with the means to find your own "fix". In my opinion, this means a useable copy of the program.

But, you may be stuck because your company accepted this machine without the proper documentation (the program) in the first place... (n)

Personally, if I were dealing with an OEM like this, it would be my LAST dealing with them.

Once again... Good luck!... :D

beerchug

-Eric
 
Haha, JRW... I know you're being facetious, but in reality, you would probably have an easier time than Ron's having...

If that 286 needs a spare part which is no longer available, they will probably be able to sell you a whole new computer for less money than that 1 spare part would cost if it were available! Remember how much that lowly 286 cost when it was "state of the art"?... :eek:

I can remember paying upwards of US$1000 for a 20meg (that MEG, not GIG!) hard drive, and that was CHEAP at the time!... :rolleyes:

beerchug

-Eric
 
Eric, glad you made that comment, we havent been dealing with the company that actually built the machine but the people that did the control/electrical aspect.

Gave me some food for thought.

Jiri the alternative was to use potentiometers for 3 of the 4 analog outputs and a board (for analog 4) to ramp the voltage reference on dancer to allow for a softstart (originally we were using a pot for that too but it needed to ramp the reference voltage so we threw something together). Technically I wont know how well it all works until I get the encoder and install it tomorrow.

Originally this machine was in another plant and every day I find more paperwork on it, I may find a hard copy of the code before its over with. I have found all the settings for the PID on the drive but that can only be a kinda reference if I change drives.

Its amazing what you can come up with if you look at the drawings long enough.

Again thanks for the comments/replies. I am building up quite a bit of paperwork to use as a reference and send/show to a few people for assistance and/or bid on the job. Reminds me I need to find my digital camera to take pictures.

Guess what, I have another rewinder to work on after I get this one going...life goes on.

Thank you all.
 
Just an update. This rewinder is working GREAT now but Wednesday morning we will be disconnecting it to replace it with an older system.

At this time I am looking at options to rebuild it (I like the way this is designed, stand alone and mobile). I am sending out specs for this machine to people to see if they want to bid on it and so are the engineers.

BUT

I am dying to do this machine myself even though I lack the experience as a programmer or drive setup tech. This is a personal accomplishment thing to me, I work with these devices daily and KNOW the components involved.

Time will tell...if I do get to do it I promise not to ask questions like "What is PID and how do you use it".

AGAIN Thanks for the ideas and thoughts.
 
rsdoran said:
This rewinder is working GREAT now but Wednesday morning we will be disconnecting it to replace it with an older system
[SARCASM]Of course, that would be the most sensible thing to do... :rolleyes: [/SARCASM]

Sound like upper management is "doing their job" again... :p

beerchug

-Eric
 

Similar Topics

I might be overthinking my problem, but I need some help. I am trying to write a formula in my plc to calculate the footage of paper being...
Replies
6
Views
581
Multiple PLCs in our plant communicate using either MSG instructions or Produce/Consume. Is there an industry best practice for documenting this...
Replies
3
Views
743
Hi Guys, Can anyone help with an easy problem? I have retrieved a prog from a faulty Mitsi FX1S plc. I managed to open the file by using a...
Replies
5
Views
2,003
Hi Team, I am working on one Paper Coating machine, My Winder, infeed, Coater and Unwinder is runnning through ABB's ACS550 VFD, My winder is...
Replies
0
Views
1,287
When I build something involving a PLC, I always like to make a paper printout. To save in a little booklet. When creating an HMI program In...
Replies
1
Views
2,034
Back
Top Bottom